Flexible and robust estimation and inference of generalised autoregressive conditional heteroscedasticity (GARCH) models with covariates ('X') based on the results by Francq and Thieu (2018) . Coefficients can straightforwardly be set to zero by omission, and quasi maximum likelihood methods ensure estimates are generally consistent and inference valid, even when the standardised innovations are non-normal and/or dependent over time, see for an overview of the package.